MLASDO Summary

Feature Importance

Importance Curve

## Warning in instance$preRenderHook(instance): It seems your data is too big for client-side DataTables. You may consider server-side
## processing: https://rstudio.github.io/DT/server.html

Random Forest Performance

## Random Forest training was skipped or metrics were not stored for this run.

SVM Results

Balanced Accuracy

## Positive class: Case 
## Negative class: Control

Mean Balanced Accuracy per kernel

kernel	Balanced_Accuracy
linear	0.5983
radial	0.6180
polynomial	0.6030

Best Hyperparameters

Distances best SVM

Anomalous Samples

Summary

## Method: Mean ± 2.00 SD 
## Positive class: Case 
## Negative class: Control

Anomalous individuals per class

	diagnosis_true	anomalous
Control	Control	4
Case	Case	9

Plot

Group Interpretability

In this section we compare clinical distributions for:

• Clinical Numeric Plot (Anomalous Negative): numeric variables between Anomalous Negative vs. Negative and Anomalous Negative vs. Positive.
• Clinical Numeric Plot (Anomalous Positive): numeric variables between Anomalous Positive vs. Positive and Anomalous Positive vs. Negative.
• Clinical Categorical Plot (Anomalous Negative): categorical variables between Anomalous Negative vs. Negative.
• Clinical Categorical Plot (Anomalous Positive): categorical variables between Anomalous Positive vs. Positive.

Clinical Numeric Plot (Anomalous Negative)

Clinical Numeric Plot (Anomalous Positive)

Clinical Categorical Plot (Anomalous Negative)

Clinical Categorical Plot (Anomalous Positive)

Individual Interpretability

Domain Expert Knowledge Table

This table summarises numeric clinical covariates with significant differences (Wilcoxon test) between the Positive and Negative classes, defines thresholds halfway between their medians, and serves as the basis to score each participant individually.

Anomalous Individuals

×